Techniques for Reducing Noise While Preserving Contrast in an Image

• US 20110090240A1
• Filed: 06/05/2009
• Published: 04/21/2011
• Est. Priority Date: 06/06/2008
• Status: Active Grant

• Current Assignee: FotoNation Limited (Adeia Inc.)
• Original Assignee: FotoNation Limited (Adeia Inc.)
• Inventors: Cohen, Noy
• Granted Patent: US 8,723,879 B2
• Time in Patent Office: Days
• US Class Current: 345/589
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: G06T 5/70 Denoising; Smoothing

Abstract

A system and method for reducing noise in images is disclosed. The present invention reduces noise and preserves contrast of an image to be displayed, the image having pixels, by (1) comparing a value of a first pixel to values of a set of other N pixels; (2) comparing values of pixels neighboring the first pixel to values of further pixels neighboring the set of other pixels; (3) determining, for each pixel in the set of other pixels, a weight based on results of steps (1) and (2); (4) calculating a de-noised pixel value based on the weights of each pixel in the set of other pixels; and (5) replacing the value of the first pixel with the de-noised pixel value.

Claims

1. A method, implemented in an image processor apparatus, for reducing noise and preserving contrast of an image to be displayed, the image having pixels, comprising:

(a) comparing a value of a first pixel to values of a set of other pixels, wherein the first pixel and the set of other pixels are each of a first color;

(b) comparing values of pixels neighboring the first pixel to values of further pixels neighboring the set of other pixels, wherein the pixels neighboring the first pixel and the further pixels neighboring the set of other pixels are one of the first color, a second color, and a third color;

(c) determining, for each pixel in the set of other pixels, a weight based on results of steps (a) and (b);

(d) calculating a de-noised pixel value based on the weights of each pixel in the set of other pixels; and

(e) replacing the value of the first pixel with the de-noised pixel value.

2. The method of claim 1, wherein step (d) comprises:

summing the weights of each pixel in the set of other pixels;

calculating a lowest power of two that is greater than the sum of the weights; and

calculating the de-noised pixel value as a function of the lowest power of two and the sum of the weights.

3. The method of claim 2, wherein the lowest power of two is adjusted prior to calculating the de-noised pixel value if a difference between the lowest power of two and the sum of the weights is lower than a predetermined value.

4. The method of claim 2, wherein the lowest power of two is adjusted prior to calculating the de-noised pixel value if the lowest power of two is lower than a predetermined value.

5. The method of claim 2, wherein calculating the de-noised pixel value comprises:

calculating a weighted average of the weights of each pixel in the set of other pixels and the weight of the first pixel, wherein the weight of the first pixel is based on the lowest power of two and the sum of the weights; and

setting the de-noised pixel value equal to the weighted average.

6. The method of claim 5, wherein the weighted average is calculated by shifting bits in a register.

7. The method of claim 1, wherein step (a) comprises:

calculating, for each pixel in the set of other pixels, a difference between the value of the first pixel and the value of each pixel in the set of other pixels.

8. The method of claim 7, wherein step (b) comprises:

calculating, for each pixel neighboring the first pixel, differences between the value of each pixel neighboring the first pixel and values of the corresponding further pixels neighboring the set of other pixels, wherein locations, relative to the first pixel, of pixels neighboring the first pixel and locations, relative to the set of other pixels, of further pixels neighboring the set of other pixels determines which pixel neighboring the first pixel corresponds to which further pixels neighboring the set of other pixels.

9. The method of claim 8, wherein the calculated differences of steps (a) and (b) are based on a spatial weight matrix.

10. The method of claim 8, wherein step (c) comprises:

adjusting a threshold value, wherein the adjustment is based on a signal level, and wherein the signal level is dependent on at least one of the value of the first pixel and the values of the set of other pixels;

comparing the adjusted threshold value and the calculated differences of steps (a) and (b); and

determining the weight of each pixel in the set of other pixels based on the comparison.

11. The method of claim 1, wherein the image comprises a detail-rich area, and wherein the detail-rich area includes the first pixel and the set of other pixels.

12. The method of claim 11, further comprising:

(f) repeating steps (a) through (e) for all pixels in the detail-rich area of the image and for all pixels near and on edges and contours in the image.

13. The method of claim 12, further comprising:

(g) outputting the image for further processing by the image processor apparatus.

14. A system for reducing noise and preserving contrast of an image to be displayed, the image having pixels, comprising:

a distance calculator that compares a value of a first pixel with values of a set of other pixels, wherein the first pixel and the set of other pixels are each of a first color, and that compares values of pixels neighboring the first pixel with values of further pixels neighboring the set of other pixels, wherein the pixels neighboring the first pixel and the further pixels neighboring the set of other pixels are one of the first color, a second color, and a third color;

a weight calculator that determines, for each pixel in the set of other pixels, a weight based on results of the distance comparator; and

a de-noised pixel calculator that calculates a de-noised pixel value based on the weights of each pixel in the set of other pixels, and that replaces the value of the first pixel with the de-noised pixel value.

15. The system of claim 14, wherein the de-noised pixel calculator comprises:

a weight adder that sums the weights of each pixel in the set of other pixels;

a minimum power-of-2 calculator that calculates a lowest power of two that is greater than the sum of the weights; and

a weighted average calculator that calculates the de-noised pixel value as a function of the lowest power of two and the sum of the weights.

16. The system of claim 15, wherein the de-noised pixel calculator further comprises a middle weight calculator that adjusts the lowest power of two prior to the weighted average calculator calculating the de-noised pixel value if a difference between the lowest power of two and the sum of the weights is lower than a predetermined value.

17. The system of claim 15, wherein the de-noised pixel calculator further comprises a middle weight calculator that adjusts the lowest power of two prior to the weighted average calculator calculating the de-noised pixel value if the lowest power of two is lower than a predetermined value.

18. The system of claim 15, wherein the weighted average calculator calculates a weighted average of the weights of each pixel in the set of other pixels and the weight of the first pixel, wherein the weight of the first pixel is based on the lowest power of two and the sum of the weights, and sets the de-noised pixel value equal to the weighted average.

19. The system of claim 18, wherein the weighted average calculator calculates the weighted average by shifting bits in a register.

20. The system of claim 14, wherein the distance calculator calculates, for each pixel in the set of other pixels, a difference between the value of the first pixel and the value of each pixel in the set of other pixels.

21. The system of claim 20, wherein the distance calculator calculates, for each pixel neighboring the first pixel, differences between the value of each pixel neighboring the first pixel and values of the corresponding further pixels neighboring the set of other pixels, wherein locations, relative to the first pixel, of pixels neighboring the first pixel and locations, relative to the set of other pixels, of further pixels neighboring the set of other pixels determines which pixel neighboring the first pixel corresponds to which further pixels neighboring the set of other pixels.

22. The system of claim 21, wherein the differences calculated by the distance calculator are based on a spatial weight matrix.

23. The system of claim 21, wherein the weight calculator comprises:

a barrel shifter that adjusts a threshold value, wherein the adjustment is based on a signal level, and wherein the signal level is dependent on at least one of the value of the first pixel and the values of the set of other pixels;

a comparator that compares the adjusted threshold value and differences calculated by the distance calculator;

and multiplexor logic that determines the weight of each pixel in the set of other pixels based on a result of the comparator.

24. The system of claim 14, wherein the image comprises a detail-rich area, and wherein the detail-rich area includes the first pixel and the set of other pixels.

25. The system of claim 24, wherein the distance calculator, the weight calculator, and the de-noised pixel calculator evaluate all pixels in the detail-rich area of the image and all pixels near and on edges and contours in the image.

26. The system of claim 25, wherein the de-noised pixel calculator outputs the image for further processing.

27. A computer program product for use with a computing device, the computer program product comprising:

a tangible computer usable medium, having computer readable program code embodied thereon, which when executed by the computing device cause the computing device to implement an image processor that reduces noise and preserves contrast of an image to be displayed, the image having pixels, the computer readable program code comprising;

first computer readable program code for causing the computing device to implement a distance calculator that compares a value of a first pixel with values of a set of other pixels, wherein the first pixel and the set of other pixels are each of a first color, and that compares values of pixels neighboring the first pixel with values of further pixels neighboring the set of other pixels, wherein the pixels neighboring the first pixel and the further pixels neighboring the set of other pixels are one of the first color, a second color, and a third color;

second computer readable program code for causing the computing device to implement a weight calculator that determines, for each pixel in the set of other pixels, a weight based on results of the distance comparator; and

third computer readable program code for causing the computing device to implement a de-noised pixel calculator that calculates a de-noised pixel value based on the weights of each pixel in the set of other pixels, and that replaces the value of the first pixel with the de-noised pixel value.

28. The computer program product of claim 27, wherein the de-noised pixel calculator:

sums the weights of each pixel in the set of other pixels;

calculates a lowest power of two that is greater than the sum of the weights; and

calculates the de-noised pixel value as a function of the lowest power of two and the sum of the weights.

29. The computer program product of claim 28, wherein the de-noised pixel calculator further adjusts the lowest power of two prior to the weighted average calculator calculating the de-noised pixel value if a difference between the lowest power of two and the sum of the weights is lower than a predetermined value.

30. The computer program product of claim 28, wherein the de-noised pixel calculator calculates the de-noised pixel value by calculating a weighted average of the weights of each pixel in the set of other pixels and the weight of the first pixel, wherein the weight of the first pixel is based on the lowest power of two and the sum of the weights, and sets the de-noised pixel value equal to the weighted average.

31. The computer program product of claim 28, wherein the de-noised pixel calculator calculates the de-noised pixel value by calculating a weighted average of the weights of each pixel in the set of other pixels and the weight of the first pixel, wherein the weight of the first pixel is based on the lowest power of two and the sum of the weights, and sets the de-noised pixel value equal to the weighted average.

33. The computer program product of claim 27, wherein the distance calculator calculates, for each pixel in the set of other pixels, a difference between the value of the first pixel and the value of each pixel in the set of other pixels.

34. The computer program product of claim 33, wherein the distance calculator calculates, for each pixel neighboring the first pixel, differences between the value of each pixel neighboring the first pixel and values of the corresponding further pixels neighboring the set of other pixels, wherein locations, relative to the first pixel, of pixels neighboring the first pixel and locations, relative to the set of other pixels, of further pixels neighboring the set of other pixels determines which pixel neighboring the first pixel corresponds to which further pixels neighboring the set of other pixels.

35. The computer program product of claim 34, wherein the differences calculated by the distance calculator are based on a spatial weight matrix.

36. The computer program product of claim 34, wherein the weight calculator:

adjusts a threshold value, wherein the adjustment is based on a signal level, and wherein the signal level is dependent on at least one of the value of the first pixel and the values of the set of other pixels;

compares the adjusted threshold value and differences calculated by the distance calculator; and

determines the weight of each pixel in the set of other pixels based on a result of the comparator.

37. The computer program product of claim 27, wherein the image comprises a detail-rich area, and wherein the detail-rich area includes the first pixel and the set of other pixels.

38. The computer program product of claim 37, wherein the distance calculator, the weight calculator, and the de-noised pixel calculator each evaluate all pixels in the detail-rich area of the image and all pixels near and on edges and contours in the image.

39. The computer program product of claim 38, wherein the de-noised pixel calculator outputs the image for further processing.

32. (canceled)